One example of a method of calculating a sound wave propagation time is recited in Patent Literature 1. In the conventional sound wave propagation time calculating method, with an M sequence phase modulated wave used as a transmission wave, an M sequence phase modulated wave transmitted by a transmitter is received by a receiver and the received signal and a transmission signal are correlated by a matched filter to detect a peak of an output of the matched filter, thereby calculating a propagation time of the sound wave by a peak time.
In addition, recited in Patent Literature 2 is one example of a method of reducing erroneous detection of a correlation peak caused by distortion of a sound wave at transmission and reception when an M sequence phase modulated wave is used as a transmission wave. In the convention method, with respect to a non-M sequence of a reception wave generated due to distortion of a sound wave transmitted or received, generating a signal whose amplitude of a half-cycle of a sine-wave pulse immediately precedent to a phase inversion part where the phase of the M sequence phase modulated signal is inverted by 180 degrees is set to be zero and driving a speaker to reduce distortion of a sound wave leads to suppression of a secondary peak generated due of non-M sequence.
One example of a method of measuring a propagation time of ultrasonic to detect a position is recited in Patent Literature 3. In the conventional position detection method using a propagation time of ultrasonic, an infrared trigger signal for a measurement start by an electromagnetic wave and an ultrasonic signal of the same waveform in each cycle are transmitted from an electronic pen simultaneously in a fixed cycle. The reception side measures a time from the reception of the trigger signal until the reception of the ultrasonic signal as a propagation time of the ultrasonic to determine a position of the electronic pen by using the propagation time.
Patent Literature 1: U.S. Pat. No. 3,876,370
Patent Literature 2: Patent Laying-Open No. 2002-286517
Patent Literature 3: U.S. Pat. No. 6,118,205
Ultrasonic signal will have a reflected wave generated because it might reflect on a surrounding wall or the like to arrive at a reception side through a plurality of paths whose propagation distance is different. Since the conventional ultrasonic propagation time measuring method has the same waveform in each transmission cycle, when a reception side receives a composite wave with a direct wave and a reflected wave of an ultrasonic signal alter receiving a trigger signal, it is impossible to precisely extract only a direct wave because discrimination between them is difficult. In particular, a configuration of the composite wave might be changed depending on how the direct wave and the reflected wave overlap with each other to make it difficult to detect an arrival point of a first direct wave.
The conventional sound wave propagation time calculating method using an M sequence phase modulated wave has a problem that even with the addition of the method of reducing distortion of a sound wave caused by transmission/reception, when a transmission source having frequent residual oscillations is used, the method is not enough to reduce distortion of a sound wave and it is accordingly difficult to suppress a secondary correlation peak generated due to non-M sequence. The reason is that even when the transmission source is driven by a signal generated whose amplitude of a half-cycle of a sine-wave pulse immediately precedent to a phase inversion part where the phase of the M sequence phase modulated signal is inverted by 180 degrees is zero, residual oscillation is generated to interfere with oscillation by a driving signal, so that a non-M sequence remains.